fabrication of hybrid thin film based on bacterial

Contents lists available at ScienceDirect

Carbohydrate Polymers

journal homepage: www.elsevier.com/locate/carbpol

Fabrication of hybrid thin film based on bacterial cellulose nanocrystals andmetal nanoparticles with hydrogen sulfide gas sensor abilityPongpat Sukhavattanakula,b, Hathaikarn Manuspiyaa,b,*a The Petroleum and Petrochemical College, Chulalongkorn University, Soi Chula 12, Phyathai Road, Pathumwan, Bangkok, 10330, Thailandb Center of Excellence on Petrochemicals and Materials Technology, Chulalongkorn University, Bangkok, 10330, Thailand

A R T I C L E I N F O

Keywords:Bacterial cellulose nanocrystalsAcid hydrolysisSilver nanoparticlesMolybdenum trioxide nanoparticlesHybrid material filmHydrogen sulfide gas sensor

A B S T R A C T

The nanocrystalline structures of bacterial cellulose (BC) are described as “environmentally friendly green na-nomaterials”. Bacterial cellulose (BC) was produced from Gluconacetobacter xylinus in pellicle form with a largebundle of fibers were acid hydrolyzed to obtain bacterial cellulose nanocrystals (BCNCs). The H2SO4 acid-hy-drolyzed BCNCs were evaluated for their smallest crystallite size and hydrodynamic size, highly negative ζ-potential value, and the highest specific surface area to interact with metallic nanoparticles. Hybrid thin film ofBCNCs based surface-loaded silver nanoparticles (AgNPs) and alginate-molybdenum trioxide nanoparticles(MoO3NPs) was developed for hydrogen sulfide (H2S) gas sensor. Sensor characteristics were investigated as wellas its response with H2S gas. The film was successfully detected H2S gas. The color of the film changed by theshift of oxidation number of MoO3NPs. Once activated by AgNPs, MoO3NPs was readily reduced to a coloredsub-oxide by atomic hydrogen that produced and received from reaction of H2S gas.

1. Introduction

Almost ten years of advances in natural-based or bio-based polymerresearch have proved the potential and importance of biopolymers for awide variety of applications, specifically for biopolymers formed bymicroorganisms or microbs, including bacterial cellulose nanocrystals(BCNCs) (Jozala et al., 2016; Lin, Huang, & Dufresne, 2012;Ummartyotin & Manuspiya, 2015). Apart from plant-derived cellulose,certain microbial-derived cellulose or bacterial cellulose produced byGluconacetobacter xylinus, tunicate, and algae are also known to producecellulose in considerable quantities in a relatively pure form eventhough the chemical structure is similar (George, Ramana, Bawa, &Siddaramaiah, 2011; Shah, Islam, Khattak, & Park, 2013). Bacterialcellulose (BC) has unique properties because of its high purity (nohemicellulose or lignin), higher surface area as compared to nativecellulose, high crystallinity, high water swelling, moldability, goodshape retention, excellent mechanical and thermal properties(Ashjaran, Yazdanshenas, Rashidi, Khajavi, & Rezaee, 2013; Corrêa, deMorais Teixeira, Pessan, & Mattoso, 2010; Fu, Zhang, & Yang, 2013;George et al., 2011; Jozala et al., 2016; Lin & Dufresne, 2014; Shahet al., 2013; Singhsa, Narain, & Manuspiya, 2018; Tang, Sisler,Grishkewich, & Tam, 2017; Vasconcelos et al., 2017).

In the area of nanomaterials technology, the top-down approaches

(e.g.: homogenization, hydrolysis, combined chemical-mechanicalprocesses) can be applied to downsize the plant-derived cellulose ormicrobial-derived fibers in small particles as cellulose nanocrystals(CNCs) suspensions, expanding versatility and new nanomaterials re-sources to this cellulosic material (Lin et al., 2012; Vasconcelos et al.,2017). These CNCs have unique nano-porous three-dimensional net-work at nanometric scales (Fu et al., 2013), and they dominates variousattractive characteristics, such as large surface area, easy surfacemodification, and high aspect ratios which can facilitate the penetrationof inorganic nanoparticles or metallic ions into the CNCs structure.

Acid hydrolysis, especially in the strong inorganic acids, such asH2SO4 and HCl are the most commonly used method for producingCNCs (Dufresne, 2012) by penetrating of hydrogen ions (H+) intoamorphous cellulose molecules promoting cleavage of covalent orglycosidic linkage/bonds, therefore releasing single crystallites(Vasconcelos et al., 2017). H2SO4 produces a highly stable colloidalsuspension because of the high negative surface charge contributed bysulfonation of the CNCs surface. However, the presence of sulfategroups (–OSO3–) reduces the thermal stability of the nanocrystals(Martínez-Sanz, Lopez-Rubio, & Lagaron, 2011). In contrast, when HClis used, a low-density surface charges is produced on the CNCs withlimited nanocrystal dispersibility, which tends to contribute sedimentin aqueous suspensions. Some studies describe the companion use of

https://doi.org/10.1016/j.carbpol.2019.115566Received 9 July 2019; Received in revised form 31 October 2019; Accepted 1 November 2019

⁎ Corresponding author at: The Petroleum and Petrochemical College, Chulalongkorn University, Soi Chula 12, Phyathai Road, Pathumwan, Bangkok, 10330,Thailand

E-mail address: [email protected] (H. Manuspiya).

Carbohydrate Polymers xxx (xxxx) xxxx

0144-8617/ © 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: Pongpat Sukhavattanakul and Hathaikarn Manuspiya, Carbohydrate Polymers, https://doi.org/10.1016/j.carbpol.2019.115566

http://www.sciencedirect.com/science/journal/01448617

https://www.elsevier.com/locate/carbpol

https://doi.org/10.1016/j.carbpol.2019.115566


mailto:[email protected]


H2SO4 and HCl to generate stable and thermally resistant CNCs sus-pensions (Corrêa et al., 2010). CNCs with different physical and me-chanical properties can be yielded by the acid hydrolysis reactionconditions, which depend on the concentration and type of acid, reac-tion time, and temperature. Many types of BCNCs have been developedfor numerous applications, including textile industry, nonwoven cloth,paper, food, waste treatment, broadcasting, mining, refineries(Ashjaran et al., 2013; Lin & Dufresne, 2014; Wu & Liu, 2012), drugdelivery systems, vascular grafts, and scaffolds for tissue engineering invitro and in vivo (Jozala et al., 2016).

Hydrogen sulfide (H2S) gas is a colorless, flammable, extremelyhazardous gas with the characteristic of a “rotten egg” smell (Chen,Wang, Hartman, & Zhou, 2008; Schirmer, Heir, & Langsrud, 2009). Itoften results from the microbial breakdown of organic materials in theabsence of oxygen. Since, it is a chemical suffocate, which could causeto lose consciousness even at a very low concentration, paralyze thelungs and affects human’s nervous system (Choudhary, Singh, Mishra, &Dwivedi, 2013; Gahlaut, Yadav, Sharan, & Singh, 2017; MaloneRubright, Pearce, & Peterson, 2017; Pandey, Kim, & Tang, 2012), and ata very high concentration (> 700 ppm) it could cause death by redu-cing the oxygen-carrying capacity of the blood, by interrupting theoxidative processes of tissue cells, and by paralyzing the nervous systemleading to respiratory failure (Chen, Morris, & Whitmore, 2013; Gahlautet al., 2017). According to the potential hazardous nature of H2S gas, itis important to detect for both human health and industrial controlpurposes, especially in the terms of raw meat products of food packa-ging industry (Sarfraz et al., 2014). Moreover, H2S gas is one of themost often encountered food spoilage gases in meat packaging systems,which is generated by the reason of microbial spoilage and lipid oxi-dation (Murphy, O’Grady, & Kerry, 2013) can change the headspace gascomposition in the package headspace (Coombs, Holman, Friend, &Hopkins, 2017; Mohebi & Marquez, 2015). Accordingly, many smartpackaging technologies based on sensor label have been exploredduring recent decades. The sensor label could be in the form of smartlabels, immobilized, or printed on/inside the packaging films, whichwould enable to detect the level of gaseous or volatile organic com-pound (VOC) content increases during the spoilage either by bacteria orenzymic degradation (Brockgreitens & Abbas, 2016; Coombs et al.,2017; McMillin, 2017; Remenant, Jaffrès, Dousset, Pilet, & Zagorec,2015; Sofos, 1994) and caution about the quality and safety of the meatproducts by changing in their color or visible spot.

Recently, the use of hybrid material, which are consisted of twoconstituents at nanometric scale. Commonly, in the organic–inorganicmatrices, these materials revealed a significant improvement in me-chanical properties of nanocomposites consisting of nanoparticles dis-persed within the matrix phase for expanding the application range ofthese materials (Gonçalves et al., 2013). For the use of inorganic ma-terials in association with almost any hybrid materials were reportedand developed for volatile substances and gasses sensor. Some recentlyreported the integration of cellulosic materials with metal oxides withthe aim of preparing a hybrid or composite allowing high accessibilityto more or less aggregated nanoparticles of incorporation with theoxide (Shimizu, Imai, Hirashima, & Tsukuma, 1999). Moreover, bac-terial cellulose, has been revealed to be a potential to offer isotropic 3Dnano-scaffold for perfect being host these inorganic compounds at thenano-scale, resulting in improvement of the inorganic phase activity,ascribed to the use of BC with high specific surface area (Foresti,Vázquez, & Boury, 2017). For example, the simplest way to achieve theintegrated BC and metal-oxide nanoparticles is to mix them undervigorous stirring (Zhang et al., 2016). Besides, a number of semi-conductors, such as MoO3 (Kabcum et al., 2016; Li et al., 2017;MalekAlaie, Jahangiri, Rashidi, HaghighiAsl, & Izadi, 2015), SnO2

(Kabcum et al., 2016; Sarala Devi, Manorama, & Rao, 1995), α-Fe2O3

(Kersen & Holappa, 2006), ZnO (Wang, Chu, & Wu, 2006), In2O3 (Xu,Wang, & Shen, 2006), WO3 (Lin, Hsu, Yang, Lee, & Yang, 1994; Solis,Saukko, Kish, Granqvist, & Lantto, 2001), and CuO (Chen et al., 2008)

have been investigated to detect and to monitor trace concentrations ofH2S. Among all reported sensor-based semiconducting metal-oxide,MoO3 based gas sensors are most widely investigated because of itssmall size, high sensitivity, fast response, simple construction, and lowcost, which has plenty of Lewis-acid sites, exhibits a high reactivity toH2S, probably because of its high chemical affinity to H2S (Kabcumet al., 2016; Li et al., 2017; MalekAlaie et al., 2015). However, a veryfast and sensitive initial response of silver nanoparticles (AgNPs) andH2S gas was very interested and reported to be a first-order reaction inAgNPs, and the initial reaction rate was proportional to the gas con-centration down to the 1 ppm volume (ppmv) level (Chen et al., 2013).Hence, the suggestion of the H2S detection using two different classesgas sensor, AgNPs fast reacted directly with H2S to generate atomichydrogen and MoO3NPs will be reduced to a colored sub-oxide byatomic hydrogen, which will become especially interesting in the im-provement of hybrid material for H2S gas sensor.

The aim of this work was to present the design of bacterial cellulosenanocrystals and metal nanoparticles based polysaccharide sodium al-ginate matrix as a hybrid thin film for H2S gas sensor. The effect of acidhydrolysis on bacterial cellulose nanocrystals preparation was in-vestigated. Preliminary experiments with H2S gas exposure at roomtemperature with time-varying have been performed. The differentconcentrations of H2S gas, especially at higher concentration, of course,reduce more Mo oxidation state, but this is out of the scope of thismanuscript.

2. Materials and methods

2.1. Materials

K. xylinus strain (TISTR No. 975) was purchased from theMicrobiological Resource Center, Thailand Institute of Scientific andTechnological Research (TISTR) (http://www.tistr.or.th/mircen/index.html). D-glucose anhydrous, yeast extract, peptone, and calcium car-bonate (CaCO3), sulfuric acid (H2SO4), hydrochloric acid (HCl; ACSreagent, 37 %), Molybdenum (VI) oxide nanopowder (MoO3NPs), so-dium borohydride (NaBH4), sodium alginate, sodium hydroxide(NaOH), and zinc chloride (ZnCl2) were obtained from Sigma-Aldrich.Silver nitrate (AgNO3) was obtained from Merck Chemicals. Sodiumsulfide nonahydrate (Na2S·9H2O,> 95 %) was purchased from AjaxFine Chem, Co, LTD, Thailand. Ultrapure water from a Milli-Q waterwas used for all of the experiments. All chemicals were used as re-ceived.

2.2. Production of bacterial cellulose (BC)

The BC hydrogel-like pellicle, biosynthesized by K. xylinus strains,was prepared by the partially modified method of (Janpetch, Saito, &Rujiravanit, 2016; Maneerung, Tokura, & Rujiravanit, 2008; Singhsaet al., 2018a). K. xylinus strains was cultured on agar containing 100 gof D-glucose anhydrous, 10 g of yeast extract, 5 g of peptone, 20 g ofCaCO3, and 25 g/L of agar at a temperature of 30 °C for 3 days. Asterilized 100mL glucose yeast extract nutrient broth (GYNB) solutionconsisted of D-glucose anhydrous 5.0.% (w/v) and yeast extract 1.0.%(w/v) in Milli-Q ultrapure water was prepared to suspend two ampoulesof the freeze-dried cells of K. xylinus (TISTR No. 975). The cell sus-pension was incubated at a temperature of 30 °C for 24 h. The 50mL ofbacteria inoculum was introduced into 2 L of GYNB and incubated at30 °C for 7 days under static conditions. After incubation, the BC pel-licle was produced on the surface of culture GYNB and then rinsed withdistilled water to remove any residual GYNB. The acquired BC pelliclewas purified by boiling in 2 % (w/v) NaOH solution at a temperature of80 °C for 1 h and then rinsed repeatedly several times with deionized(DI) water to gain the purified neutral BC pellicle.

P. Sukhavattanakul and H. Manuspiya Carbohydrate Polymers xxx (xxxx) xxxx

2

http://www.tistr.or.th/mircen/index.html

http://www.tistr.or.th/mircen/index.html

2.3. Preparation of bacterial cellulose nanocrystals (BCNCs)

The purified BC pellicles were mechanically disintegrated to a BCpaste using a laboratory blender at 5000−6000 rpm for about 15minunder ambient temperature. The BC paste was filtered through filterpaper (Whatman No. 1) to remove excess water prior to acid hydrolysis.BCNCs were prepared through acid hydrolysis. The BC paste was hy-drolyzed with acid solution in a ratio of 1:20 g/mL with continuousstirring in different conditions as follows: the 65 % H2SO4 acid solutionwith time-varying for 1, 2, and 3 h at 45 °C for H2SO4 hydrolysis by thepartially modified method of (George et al., 2011) and the 4 N HClsolution with time-varying for 4, 5, and 6 h at reflux temperature forHCl hydrolysis (designated as BCNCs-1 to BCNCs-6) (see Table S1).

The hydrolyzed BCNCs suspension from each acid hydrolysis reac-tions were then added an excess of DI water (5-fold) and cooled with anice bath to terminate acid hydrolysis reaction. The excess acidic solu-tion was removed by centrifugation at 10,000 rpm at 4 °C for 10min forprecipitating the BCNCs and then poured out the excess acidic solution.The precipitated BCNCs was collected and dialyzed (MWCO > 12,000)against DI water to gain the neutral BCNCs. After that, the neutralBCNCs was collected by centrifugation at 10,000 rpm at 4 °C for 10min.Finally, the BCNCs was removed an excess water by freeze-drying at−55 °C for 24 h to collect the dried-BCNCs.

2.4. Preparation of BCNCs based surface-loaded AgNPs (BCNCs–AgNPs)

Freeze-dried BCNCs surface were pre-loaded with silver ion (Ag+)by immersing in 1mM of the aqueous AgNO3 under ultrasonic treat-ment by using a high intensity ultrasonic bath (GT SONIC-D3) at fre-quency of 40 kHz for 3 h. After an ultrasonic treatment in AgNO3,BCNCs–Ag+ were then rinsed with ethanol for ca. 30 s to eliminate theAg+ that were unadhered and loosely adhered to the BCNCs surface.After that, the Ag+ saturated–BCNCs were ion-reduced in 100mM ofthe aqueous NaBH4 (NaBH4:AgNO3 molar ratio of 100:1) for 20min andrinsed several times (5min each) with ultra-pure water to eliminate alltraces of chemicals, by the partially modified method of (Maneerunget al., 2008). Finally, the obtained samples were frozen at -40 °C andfreeze-dried at −55 °C for 24 h to get BCNCs–AgNPs samples.

2.5. Preparation of alginate-MoO3NPs solution

A 4wt% homogeneous sodium alginate solution was prepared bydissolving sodium alginate powder in ultra-pure water at 60 °C for 4 hunder magnetic stirrer at a stirring speed of 1000 rpm. In parallel,MoO3NPs solution was prepared by dissolving required amount of100mM in ultra-pure water at room temperature for 2 h under ultra-sonic treatment by using a high intensity ultrasonic bath (GT SONIC-D3) at frequency of 40 kHz. Half the volume of the 4 wt% homogeneoussodium alginate and 100mM MoO3NPs solution were mixed at roomtemperature for 2 h under magnetic stirrer at a stirring speed of800 rpm to reach a final concentration of 2 wt% alginate–50mMMoO3NPs solution.

2.6. Preparation of hybrid BCNCs–AgNPs and alginate–MoO3NPs film

The hybrid BCNCs based surface-loaded AgNPs andalginate–MoO3NPs film was prepared by immersing 0.1 dry wt% of thefreeze-dried BCNCs-AgNPs in the 2 wt% alginate-50mM MoO3NPs so-lution at room temperature for 1 h under magnetic stirrer at a stirringspeed of 800 rpm. After then, the homogeneous BCNCs–AgNPs/alginate–MoO3NPs solution was appeared with a very light-grey color,was poured in a round-shaped poly(tetrafluoroethylene) (PTFE) or te-flon mold (10mm height; 4 mm depth; 20mm diameter), then allowedto air-dry at ambient temperature for 48 h to obtain the BCNCs–AgNPs/alginate–MoO3NPs film samples (20mm diameter; thickness 90 μm)and stored under dry conditions.

2.7. Preparation of H2S gas samples and gas exposure experiment

The ZnS powder was prepared by a simple and efficient one-potsynthesis of sodium sulfide nonahydrate (Na2S·9H2O) and zinc chloride(ZnCl2), by the partially modified method of (Infahsaeng &Ummartyotin, 2017), as follow: 1moL of ZnCl2·7H2O was prepared in500mL of DI water under magnetic stirrer at a stirring speed of 600 rpmfor 30min at temperature of 60 °C. In parallel, 1mol of Na2S·9H2O wasprepared in 500mL of DI water under magnetic stirrer at a stirringspeed of 600 rpm for 30min at 60 °C, and then added to 1mol of ZnCl2solution under magnetic stirrer at a stirring speed of 600 rpm for 1 h at60 °C. The stoichiometric of chemical reaction is

Na2S●9H2O (aq)+ ZnCl2●7H2O (aq)→ZnS (s) +Na++Cl¯

After then, the aqueous suspension of ZnS was rinsed for 10minwith methanol to eliminate all traces of chemicals, centrifuged at10,000 rpm for 20min at 4 °C to obtain the sediment ZnS, and then keptin oven at 100 °C for 24 h to obtain the ZnS powder. H2S gas was pre-pared at ambient temperature (25 °C) by the reaction of 50mM ZnSpowder with an aqueous solution of 100mM HCl (in molar ratio 1:2) ina 500mL tightly septum-capped glass bottle, which was flushed withnitrogen gas and stored at room temperature for 1 h. The excess HClsolution could prevent diffusion of generated H2S into the reactionsolution (Carpenter, Rosolina, & Xue, 2017). The stoichiometric ofchemical reaction is

ZnS (s) + 2HCl (aq)→H2S (s)+ ZnCl2 (aq)

The prepared H2S gas was calculated the mass and conversed in aconcentration of ppm by following Eqs. (1)–(3):

PV=NkBT (1)

n=N/NA (2)

m=nMA (3)

Where P is the standard ambient pressure (1.01325×105 Pa), V is thevolume of gas container (5×10-4 M3), N is the gas molecules, kB is theBoltzmann’s constant (1.38×10-23 J K-1), T is the standard ambienttemperature (298.15 K for 25 °C), n is a number of H2S moles, NA isAvogadro’s number (6.023× 1023), MA is the molecular weight of H2Sgas (34.08 gmol-1), m is the mass of H2S gas. Standard temperature andpressure (STP) is defined as a condition of 100.00 kPa (1 bar) and273.15 K (0 °C), which is a standard of the International Union of Pureand Applied Chemistry (IUPAC). Thus, the concentration of H2S gas wascalculated to be approximately 1400 ppm in the 500mL tightly septum-capped glass bottle. The H2S gas was diluted 100 times by displaced5mL of the H2S gas using syringe plunger to each of the film samplebottles with the same bottle volume. Finally, each of the film samplebottles was contained the H2S gas in concentration of approximately14 ppm. For the H2S gas exposure of BCNCs–AgNPs/alginate–MoO3NPsfilm samples. Prior to gas exposure, the film samples were prepared fordifferent gas exposure time (5min, 1 h, and 24 h) by placing in differentof the 500mL septum-capped glass bottle with an identifying mark. Thefilm samples in the septum-capped glass bottle with an identifying markwere subsequently exposed to the H2S gas, as follow: insert a cleanneedle of the syringe into the septum at the top of the bottle cap, the gaspressure in the bottle was displaced into the syringe plunger, waited forthe gas volume to reach 5mL, and then transferred the gas to each ofthe film sample bottles.

2.8. Characterization

BC production was reported as the dry weight of cellulosic fiberswithin the volume of bacterial culture media in liter (g/L). The dryweight of BC pellicle was obtained by freezing at -40 °C for 24 h prior tofreeze-drying technique at −55 °C for 24 h. The chemical structures of


3

all samples were investigated by using an attenuated total reflectancefourier transform infrared (ATR-FTIR, Nicolet Nexus 670 FT-IR spec-trometer). All spectra were scanned between 4000 and 650 cm−1 at aresolution of 32 cm−1. OPUS 5.5 software (Bruker Optics GmbH,Ettlingen, Germany) was used for IR spectral acquisition. The Iα fractionof the bacterial cellulose samples were calculated by the followingequations obtained from (Bi et al., 2014; Singhsa et al., 2018a; Singhsa,Narain, & Manuspiya, 2018), equation parameters proposed byYamamoto, Horii, and Hirai (1996).:

fα= 2.55 fαIR − 0.32 (4)

where fαIR can be calculated as Aα /(Aα + Aβ) and Aα and Aβ are theintegrated intensities from the absorbencies at corresponding wave-number 750 and 710 cm−1, respectively.

XRD diagrams of dried pristine BC and dried BCNCs samples wererecorded using a Rigaku model SmartLab 4800 diffractometer with a CuKα radiation wavelength (λ =1.54 Å), generated at a voltage of 40 kVand a filament emission of 30mA. Samples were scanned from the2θ=5 to 80° range at a scan speed of 2°/min and a scan step of 0.02°.The average crystallite size (Lhkl) were calculated based on XRD mea-surements. Lhkl was calculated from the following the Scherrer equation(Eq. (5)):

Lhkl = Kλ/βhklcosθhkl (5)

Where βhkl is the breadth of the peak of a specific phase (hkl) or is thefull width at half maximum height (FWHM) for the diffraction angle 2θin radian, K is a constant that varies with the method of taking thebreadth (K= 0.94), λ is the wavelength of incident X-rays (1.54 Å), θ isthe center angle of the peak (Bragg’s angle). The crystallinity (%) wascalculated from the following Eq. (6):

Cr(%)= (Sc/St)× 100 (6)

Where Sc is sum of net area and St is sum of total area.The ζ potentials of all acid hydrolyzed-BC suspensions were char-

acterized at 25 °C by Malvern Zetasizer Nano ZSP (Malvern InstrumentsLtd., GB).

The Z-discriminant function (Z-value) was calculated using thefollowing Eq. (7) proposed by Wada, Okano, and Sugiyama (2001):

Z=1693d1 – 902d2 – 549 (7)

where d1 is the d-spacing of the peaks at 2θ near 15° (Iα (100) and Iβ(110)) and d2 is the d-spacing of the peaks at 2θ near 17° (Iα (010) andIβ (1–10)). The relative content of cellulose Iα/cellulose Iβ ratio wascalculated using the following Eq. (8) proposed by Wada, Kondo, andOkano (2003):

fβX-ray =−70.542d1 +37.583d2 +23.360 (8)

The hydrodynamic sizes of all acid hydrolyzed-BC suspensions wereintroduced into the viewing chamber using the integrated fluidicscapability by NanoSight NS500 (Malvern Instruments Ltd., GB) by theDLS technique. The reported value is an average of 3 measurements.The total specific surface areas and micro-pore volume of selectedsamples were characterized by N2 adsorption at the temperature of li-quid nitrogen applying the Brunauer-Emmett-Teller (BET) technique(Brunauer, Emmett, & Teller, 1938)) (BELSORP, mini-II nitrogen ad-sorptometer, Osaka, Japan). As a primary step the samples were sub-jected to a freeze-drying process, and then evacuated under ambientconditions for 24 h.

The surface morphology of the samples were characterized by fieldemission scanning electron microscopy (FE-SEM, Hitachi S-4800model). The freeze-dried pristine BC, freeze-dried BCNCs, and freeze-dried BCNCs–AgNPs were sputter-coated with platinum in preparationfor FE-SEM imaging. A Hitachi model S-4800 FE-SEM microscope wasused operating at an accelerated voltage of 5 kV and a magnification of20× . SEM/EDX was used for the determination of elemental

composition of the freeze-dried BCNCs–AgNPs and the hybridBCNCs–AgNPs/alginate–MoO3NPs film. Transmission electron micro-scopy (TEM) micrographs of the pristine BC, BCNCs, and BCNCs–AgNPssamples were taken in a JEOL 100CX-2 transmission electron micro-scope at an accelerating voltage of 100 kV. Samples were diluted andprepared by dropping the sample suspension on a carbon-coated gridand allowed to dry, followed by staining with a 2 wt % aqueous uranylacetate solution. Chemical compositions, oxidation state number ofMoO3NPs on the hybrid BCNCs–AgNPs/alginate–MoO3NPs film werecharacterized by x-ray photoelectron spectroscopy (XPS, Kratos AxisUltra DLD) and with a monochromatic Al Kα as an X-ray source (anodeHT =15 kV) and energy dispersive X-ray analysis (EDX, JSM-7610 F,JEOL). The depth of penetration of XPS is 10 nm.

The binding energies were referenced to the hydrocarbon C 1s peakat 284.6 eV (Susi, Pichler, & Ayala, 2015). Data evaluation was per-formed using CasaXPS. For fitting of the Mo 5d states, we employed thefollowing parameters (Borgschulte et al., 2017; Ji et al., 2013):

Mo4+ : EB(5/2)= 229.8 eV; EB(3/2)= 233.1 eV

Mo5+ : EB(5/2)= 231.1 eV; EB(3/2)= 234.3 eV

Mo6+ : EB(5/2)= 232.8 eV; EB(3/2)= 235.9 eV

The thickness of each film sample was measured with a digitalvernier caliper (Keiba, Japan). Measurements were taken at four dif-ferent places on the film and an average value were calculated.

3. Results and discussion

3.1. Production of BC, preparation of BCNCs and their properties

For BC production. K. xylinus (or G. xylinus) strains is the originalmicrobial producer of BCNCs and has become more interesting in themodel system for the study and research of biosynthetic mechanisms ofBNC in bacteria (Keshk, 2014), because it produces a nearly nanofi-brilar film with a gelatinous layer on the opposite side and a denserdiagonal surface (Cai & Kim, 2010; Kurosumi, Sasaki, Yamashita, &Nakamura, 2009). For this study, K. xylinus strain (TISTR No. 975) wascultured on agar, introduced into 2 L of glucose yeast extract nutrientbroth (GYNB), and incubated at 30 °C for 7 days under static conditions.The average yield of freeze-dried BC produced by fermenting with D-glucose as carbon source was 2.25 g/L of culture. The physical ap-pearance of BC pellicle is white-transparent single layer with dense andsmooth surface (Fig. 1a–b) and is thus described by a highly crystallinelinear polymer of glucose synthesized, absent of hemicellulose, pectin,lignin, or any other compound found in the plant-derived cellulose (Fuet al., 2013).

BCNCs are the crystalline domains hydrolyzed through acid hy-drolysis, this method starts with the random cleavage of amorphousphases of very long cellulose micro- fibrils eventually giving raise toseveral fractions of BC hydrolysis products (George et al., 2011). Theyare quite rigid, several rod-like particles with an average length of up tohundreds of nanometers (Tang et al., 2017) and width of several nan-ometers (10−50 nm). For preparation of BCNCs, HCl and H2SO4 were

Fig. 1. BC pellicle is formed on the surface of media (GYNB) after incubated at30 °C for 7 days under static conditions (a), purified neutral BC pellicle (b),Suspension of BCNCs-3 (c) and BCNCs-6 (d).


4

used to obtain BCNCs. Six acid hydrolysis conditions with time-varyingwere used to evaluate the influence of the acid types and reaction timeon BCNCs preparation. H2SO4 produces a highly stable colloidal sus-pension (Fig. 1d) because of the high negative surface charge con-tributed by sulfonation of the CNCs surface. In contrast, when HCl isused (Fig. 1c) low-density surface charges is produced on the CNCs withlimited nanocrystal dispersibility, which tends to contribute sedimentin aqueous suspensions.

Some studies describe the companion use of H2SO4 and HCl togenerate stable and thermally resistant CNCs suspensions (Corrêa et al.,2010). CNCs with different physical and mechanical properties can beyielded by the acid hydrolysis reaction conditions, which depend on theconcentration and type of acid, reaction time, and temperature. Theabundant thick bundles of BC fibers, which corresponded to the FE-SEMimages and the TEM images (Figs. 2a and 3 a). In contrast with HCl-hydrolyzed BC for 6 h (Figs. 2b and 3 b) and H2SO4-hydrolyzed BC for3 h (Figs. 2c and 3 c) that show an increase of single fibers. The in-tegrated structure of BC fiber bundles was broken into separated fibersand many small fiber bundles by penetrating of hydrogen ions (H+)into amorphous cellulose molecules promoting cleavage of covalent orglycosidic linkage/bonds, therefore releasing single crystallites(Vasconcelos et al., 2017).

3.1.1. Z-discriminant function, Iα/Iβ ratio, crystallinity, and crystallite sizeThe XRD patterns of pristine BC and all acid-hydrolyzed BCNCs

showed three 2θ diffraction peaks at around 15°, 17°, and 22° for cel-lulose I polymorphs (both Iα and Iβ allomorphs), which are normallyascribed to crystallographic planes of 110, 010, 100 for Iα and 200,100, 1–10 for Iβ, respectively (Wada et al., 2001). To analyze and

classify all cellulose samples that were the algal-bacterial type (Iα-rich,Z > 0) or the cotton-ramie type (Iβ-dominant, Z < 0) (Wada et al.,2001), the Z-values (Table 1) were calculated by substituting the d1 andd2 values listed in Table S2 into the Eq. (7) and plotted in Fig. S1. Theintensity of the d1 is higher than the d2, which is a distinctive char-acteristic of typical cellulose that dominate mostly Iα phase (Lee, Gu,Kafle, Catchmark, & Kim, 2015). All samples were small differences ind-spacing by the acid hydrolysis and classified as Iα-rich type (Z-value> 0). Otherwise, the relative content of Iα/Iβ were calculated inTable S2. Comparing the Iα/Iβ ratios of pristine BC (0.87/0.13), BCNCs-3 (0.83/0.17), and BCNCs-6 (0.9/0.1), the highest of Iα/Iβ ratio wasfound in H2SO4-hydrolyzed BCNCs (BCNCs-6) and the lowest of Iα/Iβratio was found in HCl-hydrolyzed BCNCs (BCNCs-3). The differencesnotwithstanding, it was observed that the choice of acid hydrolysisconditions slightly affect the Iα/Iβ ratio of bacterial cellulose. Never-theless, the effect is not significant. Crystallinity is a main factor thatsignificantly influences the physical, structural and mechanical prop-erties of materials. Therefore, the XRD patterns obtained from HCl-hydrolyzed BC samples with hydrolysis time-varying at 4, 5, and 6 h(BCNCs-1 to 3) (Fig. 4a), XRD patterns obtained from H2SO4-hydro-lyzed BC samples with hydrolysis time-varying at 1, 2, and 3 h (BCNCs-4 to 6) (Fig. 4b), and XRD patterns obtained from pristine BC, BCNCs-3and BCNCs-6 (Fig. 4c) were used to determine the percent crystallinity(Cr (%)) and the crystallite size (Lhkl) of pristine BC and all acid-hy-drolyzed BCNCs. The results indicated that all BCNCs had a Cr (%)(85.4–90.2.%) greater than pristine BC (81.1.%) (Table 1). The higherin crystallinity after acid hydrolysis reaction was because of a rapid andcontinuous decrease of the amorphous content, as this amorphousphase is to a higher degree accessible to acid penetration. Crystallinity

Fig. 2. SEM micrographs of: non-hydrolyzed BC (a), BCNCs-3 (b), and BCNCs-6 (c).

Fig. 3. TEM micrographs of: non-hydrolyzed BC (a), BCNCs-3 (b), and BCNCs- 6 (c). Scale markers correspond to 200 nm.


5

is a main factor that significantly influences the physical, structural andmechanical properties of materials. Therefore, the XRD patterns ob-tained from HCl-hydrolyzed BC samples with hydrolysis time-varying at4, 5, and 6 h (BCNCs-1 to 3) (Fig. 4a), XRD patterns obtained from

H2SO4-hydrolyzed BC samples with hydrolysis time-varying at 1, 2, and3 h (BCNCs-4 to 6) (Fig. 4b), and XRD patterns obtained from pristineBC, BCNCs-3 and BCNCs-6 (Fig. 4c) were used to determine the percentcrystallinity (Cr (%)) and the crystallite size (Lhkl) of pristine BC and allacid-hydrolyzed BCNCs. The results indicated that all BCNCs had a Cr(%) (85.4–90.2%) greater than pristine BC (81.1.%) (Table 1). Thehigher in crystallinity after acid hydrolysis reaction was because of arapid and continuous decrease of the amorphous content, as thisamorphous phase is to a higher degree accessible to acid penetration.The comparison of the Cr (%) of BCNCs-3 and BCNCs-6 (which differedin acid types and the reaction time), significant difference was detected.For BCNCs-6 (involving a higher reaction time), a decrease in the Cr(%) and the crystallite size was observed, possibly due to the harshhydrolysis conditions, resulting in a probably change in the orientationof the BC main chains. Contrarily, BCNCs-3 showed a higher Cr (%)than the BCNCs obtained after acid hydrolysis H2SO4. From this, it wasevident that H2SO4 involved in the integration structure of BC fiberbundles significantly influence the cleavage of covalent or glycosidiclinkage/bonds, therefore generating smaller single crystallite size withhigher amorphous contents. Additionally, the crystallite size (Lhkl) ofpristine BC and all acid-hydrolyzed BCNCs were calculated from theXRD spectra by using the Scherrer equation (Singhsa et al., 2018a;Vasconcelos et al., 2017) (Table 1). The pristine BC showed the largerLhkl (6.9 nm) than their nanocrystals. Compared with the BCNC-3 andBCNC-6 conditions, the nanocrystals obtained under time-varying hy-drolysis conditions showed slight differences in the Lhkl values. The Lhklwas slightly decreased with increasing acid hydrolysis reaction time.The reduction in the Lhkl of acid-hydrolyzed BCNCs purposed that acidhydrolysis is tend to reducing the crystalline part of the original ma-terials (Castro et al., 2011; Singhsa et al., 2018a). Thus, we concludethat the Lhkl of H2SO4-hydrolyzed BCNCs with the highest acid hydro-lysis reaction time (BCNCs-6) were the smallest (4.6 nm) compared toother acid hydrolysis conditions; these results could confirm the highsurface area of BCNCs are successfully developed by H2SO4 acid hy-drolysis for increasing its dispersibility into other matrix materials as anexcellent additive material, while maintaining the unique properties ofpristine BC.

3.1.2. Hydrodynamic sizes and ζ-potentialThe physicochemical characterization of pristine BC and BCNCs

obtained through acid hydrolysis under different conditions (Table 1).For H2SO4 acid hydrolysis reaction, H2SO4 has the ability to esterify thehydroxyl (OHe) groups of cellulose with sulfate groups (eOSO3e) tooffer an acid half-ester or the so-called cellulose sulfate (sulfated cel-lulose) (Singhsa et al., 2018a), this is cause to generate negativelycharged surface nanocrystals. Accordingly, the strong acid hydrolyticactivity of H2SO4, the BCNCs prepared by H2SO4 hydrolysis had thesmallest hydrodynamic sizes, which were 868.5, 861.8, and 854 d.nmfor BCNCs-4, BCNCs-5, and BCNCs-6, respectively. Therefore, this helpsto separate the coalescence of nanocrystals forced by hydrogenbonding, and then the consistent well-dispersed BCNCs suspension can

Table 1The physicochemical characterization of pristine BC and BCNCs obtained through acid hydrolysis under different conditions.

Sample XRD FT-IR Zetasizer BET Nanosizer

Crystalsize (nm)

Crystallinity(%)

Z-values Cellulose Iα content(%)

ζ-potential(mV)

Specific surface area, As

(m2/g)Micro-pore volume, Vm

(cm3(STP)/g)Hydrodynamic size(d.nm)

Pristine BC 6.9 81.1 +8.47 90.1 n.d. 3.1 0.72 n.d.BCNCs-1 5.9 86.4 +8.63 91.0 −21.5 ± 4.7 n.d. n.d. 1153 ± 478.5BCNCs-2 5.8 88.7 +7.49 91.3 −21.9 ± 4.9 n.d. n.d. 1142 ± 444.1BCNCs-3 5.3 90.2 +7.45 91.1 −22.6 ± 4.3 6.0 1.38 1122 ± 338.7BCNCs-4 4.9 85.4 +8.59 91.8 −33.5 ± 6.6 n.d. n.d. 868.5 ± 424.4BCNCs-5 4.7 86.7 +8.99 91.0 −34.2 ± 5.3 n.d. n.d. 861.8 ± 356.6BCNCs-6 4.6 88.5 +9.17 92.3 −34.3 ± 6.4 6.3 1.45 854 ± 370.5

n.d.: not determine.

Fig. 4. XRD patterns obtained from HCl-hydrolyzed BC with hydrolysis time-varying at 4, 5, and 6 h (BCNCs-1 to 3) (a), XRD patterns obtained from H2SO4-hydrolyzed BC with hydrolysis time-varying at 1, 2, and 3 h (BCNCs-4 to 6) (b),XRD patterns obtained from non-hydrolyzed BC, BCNCs-3, and BCNCs-6 (c).


6

be received. In contrast, HCl has milder acid-hydrolysis reaction andbasically promotes cellulose crystallites with native crystalline struc-tures that have no evident of sulfate groups, resulting in the productionof neutral nanocrystals. Therefore, the hydrodynamic sizes of the HCl-hydrolyzed BCNCs were larger, 1153, 1142, and 1122 d.nm for BCNCs-1, BCNCs-2, and BCNCs-3, respectively.

The zeta potential is commonly applied to indicate the presence ofsurface charges on any material that can be correlated to the zeta po-tential of the particles. In addition, zeta potential is important for asuperior understanding on the character of the particle stability, and ithas also been well-known that suspension particles with higher absolutevalues of the zeta potential have a tendency to be less aggregated byreason of high electrical repulsion among the particles (Elimelech,Nagai, Ko, & Ryan, 2000; Vasconcelos et al., 2017). Zeta potential va-lues with a modulus with higher than 30mV reflect good stability of acolloidal suspension (Vasconcelos et al., 2017). The all tested acid-hy-drolysis reaction resulted in BCNCs suspensions with high zeta potentialvalues, especially in H2SO4, confirming their remarkable stability. Thenanocrystals had a highly negative surface-charge density because ofthe conjugated effect among sulfate groups (eOSO3e) derived fromesterification and the hydroxyl groups present on the BCNCs surface.Thus, the H2SO4-hydrolyzed BCNCs was modified with sulfate groupson their surface and became a highly negative sulfated BCNCs, resultingin the highly negative surface ζ-potential values of -33.5, -34.2, and-34.3 mV for BCNCs-4, BCNCs-5, and BCNCs-6, respectively. In contrast,HCl has milder acid-hydrolysis reaction, resulting in their ζ-potentialvalues were insignificant negative compared to H2SO4-hydrolyzedBCNCs, about -21.5, -21.9, and -22.6mV for BCNCs-1, BCNCs-2, andBCNCs-3, respectively. For HCl-hydrolyzed BCNCs, the presence oflower negative charges could be the reason of a significantly increasedof oxidation of the BC main chains by reactive oxygen speciesthroughout BC acid hydrolysis, causing negatively charged functionalgroups, such as for example, carboxylic groups on the surface of na-nocrystals (Winter et al., 2010). Consequently, the remarkable hydro-dynamic sizes and ζ-potential values of the H2SO4-hydrolyzed BCNCswas successfully developed for increasing its dispersibility on alginate-based film while maintaining the unique properties of pristine BC, thuspreventing their aggregation, and also for increasing AgNPs dis-persibility, leading AgNPs to be more possibility of access to H2S gas.

3.1.3. Total specific surface area and micro-pore volumeThe BET-adsorption isotherms curve of pristine BC, HCl-hydrolyzed

BCNCs (BCNCs-3), and H2SO4-hydrolyzed BCNCs (BCNCs-6) werecharacterized (Fig. S2) in order to identify the total specific surfacearea, As (m2/g) and micro-pore volume, Vm (cm3(STP)/g) (Table 1). Asa result of the strong acid hydrolytic activity, the BCNCs prepared byHCl and H2SO4 hydrolysis, we should expect an enlarged total specificsurface. Comparing the As of pristine BC (3.1 m2/g), BCNCs-3 (6.0m2/g), and BCNCs-6 (6.3 m2/g), the highest of As were found in H2SO4-hydrolyzed BCNCs (BCNCs-6) and the lowest of As were found in pris-tine BC. The results of this measurement are closely linked with those ofthe ζ-potential analysis because the higher ζ-potential value that pro-motes the cellulose fibers to be less aggregated in order to increase As

value. Moreover, the micro-pore volumes was investigated. Since thebacterial cellulose fibers have only rod-like or whisker-shaped particles,it was suggested that they are absolutely non-porous exhibiting onlyvery small micro-pore volumes (Bismarck et al., 2002). Comparing theVm of pristine BC (0.72 cm3(STP)/g), BCNCs-3 (1.38 cm3(STP)/g), andBCNCs-6 (1.45 cm3(STP)/g), the highest of Vm were found in H2SO4-hydrolyzed BCNCs (BCNCs-6) and the lowest of As were found in pris-tine BC. It was found that the higher value in the Vm area and As werepromoted by the de-aggregated cellulose fibers with highly negativesurface charges. It is highly recommended that H2SO4-hydrolyzedBCNCs (BCNCs-6) could be used for increasing AgNPs dispersibility.

3.1.4. Identification of cellulosic fibres and cellulose Iα content (%)The FTIR spectra of pristine BC, BCNCs-3, and BCNCs-6 (Fig. S3)

showed typical cellulose vibration bands, such as 3313 cm-1 (stretchingof OeH bonds), 1428 cm-1 (asymmetric angular deformation of CeHbonds), 1163 cm-1 (asymmetrical stretching of CeOeC glycosidebonds), 1110 cm-1 and 1059 cm-1 (stretching of CeOH and CeCeOHbonds in secondary and primary alcohols, respectively), 1024 cm-1

(CeOeC anti-symmetric stretching), 896 cm-1 (angular deformation ofCeH bonds) (Vasconcelos et al., 2017), and 700−800 cm−1 (H-bonding vibrations) (Szymańska-Chargot, Cybulska, & Zdunek, 2011).However, there was no evident IR peak of the 807 cm-1 band for CeSbond symmetric vibration of CeOeSO3e groups, generated by ester-ification of H2SO4 acid-hydrolysis reaction by the reason of the smallamount of the attached sulfate groups (Singhsa et al., 2018a). Normally,cellulosic materials are homogenous polycrystalline macromolecularcompound, and it is consisted of crystalline (main ordered) and amor-phous (less ordered) regions (Bi et al., 2014). Besides, it is recognizedthat the crystalline region of cellulose I is a compound of two distinctcrystalline types: meta-stable state celluloses Iα (triclinic) and stablestate celluloses Iβ (monoclinic) (Ross, Mayer, & Benziman, 1991). TheFT-IR spectroscopy was performed in order to identify the Iα content ofbacterial cellulose samples that could be evaluated by the integratedintensities from the IR absorbencies at around 750 and 710 cm−1 (Fig.S4). Table 1 shows cellulose Iα content (%) of pristine BC and BCNCsobtained through acid hydrolysis under different conditions, with thevalues ranged from 92.3–90.1%. Hence, it is confirmed that the BC isrich cellulose Iα contents, whereas plant cellulose is rich in cellulose Iβ(Atalla & Vanderhart, 1984). Comparing the Iα content (%) of pristineBC and all BCNCs conditions, the highest Iα content (92.3%) were foundin H2SO4-hydrolyzed BCNCs (BCNCs-6) and the lowest of Iα content(90.1%) were found in pristine BC. Furthermore, these results wereclosely related to the Iα/Iβ ratios obtained from XRD analysis. However,the Iα content (%) were not significantly different among samples indifferent conditions. It suggested that an increase of acid hydrolysisreaction time, types of acid and the esterification of sulfate groupscould not significantly influence the Iα content (%) of bacterial cellu-lose.

3.2. Preparation of BCNCs based surface-loaded AgNPs and their properties

Accordingly, the previous results in Table 1, we found that suitablecondition of acid hydrolysis reaction for preparing BCNCs with H2SO4

hydrolyzed BCNCs at 45 °C for 3 h (BCNCs-6) when compared to thosewith different conditions, results in the smallest hydrodynamic sizes(854 d.nm) the smallest crystallite sizes (4.6 nm), and the highest ne-gative surface ζ-potential values (-34.3mV); these results could confirmthe high surface area and the high dispersibility by increasingly inter-acting with other materials.

3.2.1. Formation of AgNPs on BCNCs-6By the immersion of BCNCs-6 in silver ion Ag+ aqueous solutions

using Ag(NO3) as Ag+ precursors, the high surface of the several singlefibers with three-dimensional network structure of the BCNCs-6, asshown in Fig. 2c, could allow Ag+ ions to penetrate into the BCNCs-6structure as long as the adsorption equilibrium is accomplished. Ac-cording to, Ag+ ions is impregnated and loaded onto the stable ab-sorption site via electrostatic repulsive interaction between Ag+ ionsand lone pair electrons of OH (hydroxyl) groups on the surface ofBCNCs-6 nanofibers (Azizi, Ahmad, Mahdavi, & Abdolmohammadi,2013; Janpetch et al., 2016), as shown in chemical Eqs. (9)–(12):

4Ag(NO3) (s)+ 2H2O→4Ag++4HNO3+O2 (9)

Ag++ surface of BCNCs→Ag+.. .BCNCs (10)

Ag+.. .BCNCs+OH-→AgOH.. . BCNCs (11)


7

AgOH.. . BCNCs→AgNPs.. . BCNCs+H2O (12)

After the stable absorption site via electrostatic repulsive interactionof Ag+ ions on BCNCs fibers, the Ag+ saturated-BCNCs were ion-re-duced and transformed to silver nanoparticles (AgNPs) deposited on thesurface of BCNCs fibers via the chemical reaction with OH- fromNH4OH (Janpetch et al., 2016; Li, Shi, Zhong, & Yin, 1999; Yang, Zhu,Liu, Chen, & Ma, 2008), as shown in chemical Eq. (13)

2OH–+2Ag+→AgO+AgNPs.. . BCNCs+H2O (13)

Consequently, morphological properties of the obtained dryBCNCs–AgNPs samples were analyzed.

3.2.2. Microstructural characteristics of BCNCs–AgNPsMorphological analysis of the obtained BCNCs–AgNPs samples by

TEM micrograph (Fig. 5), the average size and shape of the AgNPs wereapproximately in the range from 1 nm to 10 nm although some ag-glomeration was also present. The results from these micrographs couldconfirm the well-dispersed of AgNPs with the high surface area ofBCNCs-6 by increasingly interacting surface between each other.

3.2.3. Elemental analysis of BCNCs–AgNPsThe elemental composition of BCNCs-AgNPs was determined by the

energy-dispersive X-ray (EDX) analysis, the samples of BCNCs-AgNPswere coated with platinum (Pt) by ion sputtering. The EDX images inFig. S5a and b were analyzed by X-ray dot mapping mode and the EDXimages in Fig. S5c was taken by using element identification mode. InFig. S5c, EDX images of BCNCs–AgNPs sporadically surface-loaded byAgNPs with which they closely interact to BCNCs-6 to form a physicaladsorption. There are significant peaks of the Ag element, which belongto AgNPs is loaded on the surface of BCNCs-6 with homogenouslydispersed, as shown in Fig. S5b. Thus, it is confirmed that the hy-bridization of AgNPs and BCNCs by using the chemical reducing re-agent under ultrasonic treatment solution could accomplish the hy-bridization by physical adsorption of AgNPs on BCNCs’s surface. Theyalso exhibit the rather high peak of the C and O element, which are ageneral component of BCNCs and other cellulosic material.

3.3. Preparation of hybrid BCNCs–AgNPs/alginate–MoO3NPs film andtheir properties

3.3.1. Formation of hybrid BCNCs–AgNPs/alginate–MoO3NPs filmThe demonstration of the hybrid material of BCNCs–AgNPs/

alginate–MoO3NPs prepared by physical adsorption techniques and itsfilm form are shown in Fig. 6a and b. The hybrid of the BCNCs–AgNPs/alginate–MoO3NPs in the form of solution and film can be both in Mo(V)O3 (Mo5+) and Mo(VI)O3 (Mo6+) characterized by x-ray photo-electron spectroscopy (XPS) technique.

The preparation of Mo(VI)O3NPs solution with ultrapure water Eq.(14), Mo5+●H2O and Mo6+●H2O were occurred without separation ofthe new phase. The two water molecules in Mo5+●H2O and Mo6+●H2O

Fig. 5. TEM micrographs of BCNCs-6 based surface-loaded AgNPs. TEM, scalemarkers correspond to 100 nm.

Fig. 6. Schematic demonstration of the hybrid material of BCNCs–AgNPs/ alginate–MoO3NPs prepared by physical adsorption techniques (a) and its film form (b).


8

are different bonded types, depending on the oxidation state of Mo. ForMo5+●H2O, in addition of oxygens, is coordinated to Mo atoms in onelayer to oxygen ions in the next layer, and for the Mo6+●H2O is held byhydrogen bonding in between the layers of MoO3 (Günter, 1972). Bythe immersion of sodium alginate in the Mo(VI)O3NPs solution, thecarboxylate side groups with several of the negative charges of alginateEq. (15), is highly reactive to cations of Mo5+ to form Mo5+Alginate,and also to electrostatic repulsive interaction with Mo6+ to form Mo6+..

.Alginate– Eq. (16).

Mo(VI)O3NPs (s)+ 2H2O→Mo5+●H2O+Mo6+●H2O (14)

Sodium alginate (s) +H2O→Na++Alginate- (15)

Mo5+●H2O+Mo6+●H2O+Alginate–→Mo5+Alginate+Mo6+.. .Al-ginate– (16)

By hybridizing the dried BCNCs–AgNPs to alginate–MoO3NPs so-lution, there are two different types of surface negative charges on thesurface of BCNCs–AgNPs: (i) from sulfate groups (SO3–); and (ii) fromhydroxyl groups (OH–), and thus causing different interactions with Moand alginate (Eqs. (17)–(18)).

Mo5+Alginate+Mo6+.. .Alginate-+ BCNCs–AgNPs (SO3–)→Mo5+

Alginate+Mo6+. .Alginate–+Mo5+BCNCs–AgNPs (SO3–)+Mo6+..

.BCNCs–AgNPs (SO3–) (17)

Mo5+Alginate+Mo6+.. .Alginate–+BCNCs–AgNPs (OH–)→Mo5+

Alginate+Mo6+. . .Alginate–+Mo5+BCNCs–AgNPs (OH–)+Mo6+..

.BCNCs–AgNPs (OH–) (18)

Consequently, properties of the obtained hybrid BCNCs–AgNPs/alginate–MoO3NPs film samples were analyzed and H2S gas sensorability was characterized using XPS with H2S gas exposure experiment.

3.3.2. Physical appearances of the filmThe hybrid BCNCs–AgNPs/alginate–MoO3NPs films were visually

homogeneous with no air bubbles, wrinkles, snags, or brittle areas andwere easily manipulatable and moderately flexible. The appearance ofthe film was not significantly affected by the addition of bacterial cel-lulose and metal nanoparticles at a 50mM of MoO3 solution and 0.1 drywt% of BCNCs–AgNPs except for the film color which was changed verylittle from its initial alginate film (transparent white color). Thetransparency of the film was evaluated visually and photographed(Fig. 6b), showing the development of a light greyish-white color, de-pending on the MoO3 content. The average thickness of the film mea-sured by a micrometer was approximately 90 μm.

3.3.3. Identification of alginate and MoO3NPsFTIR analysis of the BCNCs-6, BCNCs-6–AgNPs/alginate–MoO3NPs

film, and alginate film attempted to characterize the hybridization ofBCNCs-6–AgNPs and alginate–MoO3NPs by distinguishing the infrared(IR) bands and vibration shifts related to the BCNCs-alginate andMoO3NPs–alginate interactions (Fig. S6). The characteristic peak ofalginate polysaccharide vibration bands, such as the peak at 1602 cm-1

(eC]O carbonyl bonds), 1424 and 1024 cm-1 in both alginate andBCNCs were assigned to (eCOO and eCeO carboxyl stretching bands)(Deepa et al., 2016), 3313 cm-1 (eOeH aliphatic chain stretching vi-brations), and 2930 cm-1 (eCeH aliphatic chain stretching vibrations).For the hybrid BCNCs-6–AgNPs/alginate–MoO3NPs film, the Mo]Obonding vibration bands is clearly shown as a sharp band centered atabout 950 cm-1, and characteristic peaks were present in the953−957 cm-1 range, which can be determined to the stretching modeof the terminal Mo]O group (Afsharpour, Mahjoub, & Amini, 2009;Bhattacharya, Dinda, & Saha, 2015) and 865 cm-1 are assigned to theMo–O–Mo stretching vibrations of MoO3 (Bhattacharya et al., 2015).Some reports showed a high peak intensity of terminal M]O bonds(993 cm-1 and 817 cm-1) presents a large oxygen vacancy (Bhattacharyaet al., 2015). Some differences can be observed after MoO3NPs addition

into the alginate matrix. Consequently, occurrence of the peak with asharp band centered at about 950 cm-1 and 865 cm-1 were observed asthe MoO3NPs was incorporated into the alginate matrix. These resultscan represent the existence of the ionic bonding and electrostatic re-pulsive interaction between the alginate and MoO3NPs. MoO3NPs in theform of Mo(V)O3 (Mo5+) are formed an ionic bonded with carboxylicacid groups (eCOOe) on the alginate chain, and MoO3NPs in the formof Mo(VI)O3 (Mo6+) are physically adsorbed on both in the alginatechain and sulfate groups (eOSO3e) on the BCNCs’s surface by elec-trostatic interaction.

3.3.4. Elemental analysis of BCNCs–AgNPs/alginate–MoO3NPs filmThe elemental composition of the hybrid BCNCs–AgNPs/

alginate–MoO3NPs film was determined by EDX analysis, the filmsample was coated with platinum (Pt) by

ion sputtering. The EDX images in Fig. S7a–c were analyzed by X-ray dot mapping mode and the EDX images in Fig. S7d was taken byusing element identification mode. In Fig. S7d, EDX images of the hy-brid BCNCs–AgNPs/alginate–MoO3NPs film based on the alginate ma-trix, sporadically hybridized by MoO3NPs and BCNCs–AgNPs withwhich they closely interact to alginate to form a physical adsorption.There are significant peaks of the Ag element, which belong to AgNPsare loaded on the surface of BCNCs-6 with homogenously dispersed, asshown in Fig. S7b. Besides, the significant peaks of the Mo element,which belong to MoO3NPs are loaded on the alginate with homo-genously dispersed, as shown in Fig. S7c. Thus, it is confirmed that thehybridization of: (i) alginate and MoO3NPs by using ultrasonic treat-ment solution could accomplish the hybridization by physical adsorp-tion by covalent and ionic bonds of MoO3NPs with carboxylic acidgroups (eCOOe) on the alginate chain; and (ii) BCNCs–AgNPs by usingthe chemical reducing reagent under ultrasonic treatment solutioncould accomplish the hybridization by physical adsorption of AgNPs onBCNCs’s surface. They also exhibit the rather high peak of the C and Oelement, which are a general component of alginate polysaccharide,BCNCs and other cellulosic material.

3.4. H2S gas exposure experiment

3.4.1. H2S gas sensor ability of hybrid BCNCs–AgNPs/alginate–MoO3NPsfilm

For the H2S gas exposure experiment, the demonstration of the in-house H2S gas preparation and H2S gas exposure experiment of hybridBCNCs–AgNPs/alginate–MoO3NPs film is shown in Fig. S8. The con-centration of H2S gas was calculated to be approximately 1400 ppm inthe tightly septum-capped glass bottle and then was diluted 100 timesby displaced 5mL of the H2S gas using syringe plunger to each of thefilm sample bottles with the same bottle volume. Finally, each of thefilm sample bottles was reached the H2S gas in concentration of ap-proximately 14 ppm and leave them with gas exposure time-varying for5min, 1 h, and 24 h.

The fast H2S gas sensor of the hybrid BCNCs–AgNPs/alginate–MoO3NPs film would be highly useful for a low concentrationof H2S at 14 ppm, in which increase of the Mo sub-oxide color intensityaccording to the increase of H2S gas exposure time (Fig. 7a–d). Thechange in color of the hybrid BCNCs–AgNPs/alginate–MoO3NPs film

Fig. 7. H2S gas exposure of hybrid BCNCs–AgNPs/alginate–MoO3NPs film;before exposing to H2S gas (a), after exposing to 14 ppm of H2S for 5min (b),1 h (c), and 24 h (d).


9

from transparent light-grey to opaque dark brown-black can be as-signed to the conversion of Ag to Ag2S to produce atomic H, and thenthe Mo to from MoO3NPs are readily reduced to a colored sub-oxide byatomic H.

3.4.2. Mechanism of H2S sensingPristine MoO3 consists of Mo(VI)O3(Mo6+) forming the conduction

bands and O2− forming the valence bands (Borgschulte et al., 2017) canbe assigned to Mo6+ 3d5/2and Mo6+ 3d3/2, respectively (Gong & Haur,2017; Ji et al., 2013). The first reaction is believed to occur at the HS–adsorbed on the AgNPs of the well-dispersed BCNCs–AgNPs to readilyform Ag2S and promptly generated H2. Next, the intercalation of H2

into MoO3 by the vicinity of H2 to oxygen without alteration of thestructural motif apart from few increase of the cell volume (Borgschulteet al., 2017), and water is formed at longer gas exposure time. Thedissociation of hydrogen molecule from the reaction of AgNPs and H2Sgas on oxide surfaces of MoO3NPs in the film, which decomposes intoMoO2

+1/2 and H2O by the fast decrease of Mo oxidation state fromMo6+ into Mo5+, slowly forms Mo4+ states, forms a very low-intensityMoS24+ at exposure time of 1 h, and forms a low-intensity MoS24+ atexposure time of 24 h. Besides, the increase of AgNPs dispersibility bywell-dispersed BCNCs–AgNPs in alginate–MoO3 matrix phase, leadingAgNPs to be more possibility of access to H2S gas. The Mo(IV)S2(Mo4+)is considerably increased due to the loss of O2, depending on gas ex-posure time of sampling. These conversion can be described by thechemical reactions as shown in chemical Eqs. (19)–(22):

8Ag+4HS–→4Ag2S+ 2H2+ 4e– (19)

Mo(VI)O3+H2→Mo(V)O3+H2+ e– (20)

Mo(VI)O3+H2→Mo(IV)O2+H2O (21)

Mo(IV)O2+2HS–→Mo(IV)S2+H2+O22– (22)

The rate of reaction increases with the increase in gas exposuretime. These mechanism fully support the H2S gas sensor from the hy-brid BCNCs–AgNPs/alginate–MoO3NPs film by this gas exposure ex-periment.

3.4.3. XPS measurementH2S gas sensing characteristics of the films were examined from XPS

spectrogram. A high-resolution XPS measurement was carried out inorder to examine the surface chemical composition and the oxidationstates of Mo. Fig. 8 shows the XPS spectra of the hybrid BCNCs–AgNPs/alginate–MoO3NPs films in order to examine the oxidation states of theMo 3d, 5d, and 2 s region for Mo. According to the XPS spectra, all peakpositions for Mo with different oxidation state are shown in the TableS3. The peaks at 232.8 and 235.9 eV, characteristic of MoO3, can beassigned to Mo6+ 3d5/2 and Mo6+ 3d3/2, respectively (Ji et al., 2013).Under the H2S gas exposure, the peak intensity of Mo5+ 3d5/2, Mo6+

3d3/2, and Mo6+ 3d5/2 XPS spectra (Fig. 8b–d) are prone to be reducedespecially for 24 h, indicating a significant amount of Mo6+ in MoO3

has been reduced to sub-oxide state to promote the change in color ofthe film.

The low energy peak at 229.8 eV can be observed and assigned toMo4+ 3d5/2 by reason of the reduction that occurred after H2S gasexposure for 5min, and the peak intensity is considerably increased atgas exposure for 1 h and 24 h, respectively. The signals from MoO3

under gas exposure is considered to be along with the conversion thatdescribed by the chemical reactions as shown in chemical Eqs.(19)–(22).

For gas exposure at 24 h, the peak intensity at 229.8 eV (Mo4+ 3d5/2) and 226.25 eV (MoS24+ S 2 s) are much stronger than that of pristineMoO3, indicating that on the MoO3NPs surface the relative concentra-tion of Mo4+ is higher than that of MoO3 and the oxidation state of Mois lower. The peak areas of Mo4+/Mo6+ ratio from XPS spectra aredetermined to be 0.136, 0.477, and 2.037 for H2S gas exposure time for

Fig. 8. XPS spectra of the BCNCs–AgNPs/alginate–MoO3NPs films in order to examine the valence states of the Mo 3d, 5d, and 2 s region for Mob; before exposing toH2S gas (a), after exposing to 5mL of H2S gas for 5min (b), 1 h (c), and 24 h (d).


10

5min, 1 h, and 24 h, respectively as shown in Table S4.The ratio is determined based on: (i) the oxidation state or valence

state of MoO3 after losing the O molecule by H2 intercalation (Mo(IV)O2(Mo4+)); and (ii) the oxidation state of pristine MoO3 (Mo(VI)O3(Mo6+)). Thus, the conversion of Mo6+ to Mo4+ is depend on gasexposure time with the slower increase of the ratio value due to somelimitation of Mo oxide surfaces for H2 intercalation.

4. Conclusion

In this work, BC was successfully produced by bacterial strain of K.xylinus under static culture with average yield was 2.25 g/L of culture.The acid conditions profoundly influenced the physical properties ofnanocrystals including the morphology, hydrodynamic size, ζ-potential,crystallite size, and CI. Accordingly, BCNCs was successfully achievedby acid hydrolysis under various acid conditions, namely, HCl, H2SO4

with time-varying. Resulting in H2SO4 acid-hydrolyzed BCNCs (BCNCs-6) were evaluated for their highest surface area (6.3m2/g) for inter-action with metallic nanoparticles with the smallest crystallite size(4.6 nm) and hydrodynamic size (854 d.nm) with the most remarkableζ-potential value (-34.3 eV) for negative surface charge. Thus, BCNCs-6was selected for the study of BCNCs surface-loaded AgNPs and thenhybridized with alginate-MoO3NPs by physical adsorption under ul-trasonic treatment to get film structure for H2S gas sensor ability. Thus,the high surface area of BCNCs increase its dispersibility on alginate-based film while maintaining the unique properties of pristine BC, thuspreventing their aggregation, and also for increasing AgNPs dis-persibility, leading AgNPs to be more possibility of access to H2S gas.H2S gas with concentration 14 ppm, have been utilized for the mea-surements. The hybrid film of BCNCs–AgNPs/alginate–MoO3NPs wassuccessfully developed for H2S gas sensor by reason of a change in Mooxidation state and the reduction mechanism by atomic hydrogen in-tercalating of MoO3NPs on the film. The Mo oxidation state on the filmunder H2S gas exposure can be Mo(IV)O3(Mo4+), Mo(V)O3(Mo5+), andMo(VI)O3 (Mo6+) examined from XPS spectrogram. The film color arechanged from transparent light greyish-white to opaque dark brown-black indicated the dissociation of hydrogen molecule from the reactionof AgNPs and H2S gas on oxide surfaces, which decomposes intoMoO2

+1/2 and H2O by the fast decrease of Mo oxidation state fromMo6+ into Mo5+, slowly forms Mo4+ states, forms a very low-intensityMoS24+ at exposure time of 1 h, and forms a low-intensity MoS24+ atexposure time of 24 h. The color intensity is gradually increased ac-cording to the increase of exposure time. Consequently, this study couldprovide useful information about the preparation of rod-like BCNCsfrom BC produced by K. Xylinus strain under static culture, surface-loaded with metal nanoparticles, and hybrid with alginate–MoO3 toreach H2S gas sensor ability, in addition to rendering the properties ofnanocrystals in accordance with the desired applications.

Acknowledgements

This research was supported by the Petroleum and PetrochemicalCollege, Chulalongkorn University. One of the authors, P.S., would like toacknowledge the scholarship from the 100th Anniversary ChulalongkornUniversity Fund for Doctoral Scholarship and the 90th Anniversary ofChulalongkorn University Scholarship (Ratchadaphiseksomphot EndowmentFund) as well as supported by Graduate School of Chulalongkorn University.Part of research grant was supported by Center of Excellence onPetrochemicals and Materials Technology, Chulalongkorn University.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in theonline version, at doi:https://doi.org/10.1016/j.carbpol.2019.115566.

References

Afsharpour, M., Mahjoub, A., & Amini, M. M. (2009). Synthesis of molybdenum oxidenanohybrids as efficient catalysts in oxidation of alcohols. Journal of Inorganic andOrganometallic Polymers and Materials, 19.

Ashjaran, A., Yazdanshenas, M. E., Rashidi, A., Khajavi, R., & Rezaee, A. (2013).Overview of bio nanofabric from bacterial cellulose. Journal of the Textile Institute,104(2), 121–131.

Atalla, R. H., & Vanderhart, D. L. (1984). Native cellulose: A composite of two distinctcrystalline forms. Science, 223(4633), 283–285.

Azizi, S., Ahmad, M., Mahdavi, M., & Abdolmohammadi, S. (2013). Preparation, char-acterization, and antimicrobial activities of ZnO nanoparticles/cellulose nanocrystalnanocomposites. BioResources, 8.

Bhattacharya, S., Dinda, D., & Saha, S. K. (2015). Role of trap states on storage capacity ina graphene/MoO3 2D electrode material. Journal of Physics D: Applied Physics, 48.

Bi, J. C., Liu, S. X., Li, C. F., Li, J., Liu, L. X., Deng, J., ... Yang, Y. C. (2014). Morphologyand structure characterization of bacterial celluloses produced by different strains inagitated culture. Journal of Applied Microbiology, 117(5), 1305–1311.

Bismarck, A., Aranberri, I., Springer, J., Lampke, T., Wielage, B., Stamboulis, A., ...Limbach, H.-H. (2002). Surface characterization of flax, hemp and cellulose fibers;surface properties and the water uptake behavior. Polymer Composites, 23, 872–894.

Borgschulte, A., Sambalova, O., Delmelle, R., Jenatsch, S., Hany, R., & Nüesch, F. (2017).Hydrogen reduction of molybdenum oxide at room temperature. Scientific Reports, 7,40761.

Brockgreitens, J., & Abbas, A. (2016). Responsive food packaging: Recent progress andtechnological prospects. Comprehensive Reviews in Food Science and Food Safety, 15(1),3–15.

Brunauer, S., Emmett, P. H., & Teller, E. (1938). Adsorption of gases in multimolecularlayers. Journal of the American Chemical Society, 60(2), 309–319.

Cai, Z., & Kim, J. (2010). Bacterial cellulose/poly(ethylene glycol) composite:Characterization and first evaluation of biocompatibility. Cellulose, 17(1), 83–91.

Carpenter, T. S., Rosolina, S. M., & Xue, Z.-L. (2017). Quantitative, colorimetric paperprobe for hydrogen sulfide gas. Sensors and Actuators B, Chemical, 253, 846–851.

Castro, C., Zuluaga, R., Putaux, J.-L., Caro, G., Mondragon, I., & Gañán, P. (2011).Structural characterization of bacterial cellulose produced by Gluconacetobacterswingsii sp. from Colombian agroindustrial wastes. Carbohydrate Polymers, 84(1),96–102.

Chen, J., Wang, K., Hartman, L., & Zhou, W. (2008). H2S detection by vertically alignedCuO nanowire array sensors. The Journal of Physical Chemistry C, 112(41),16017–16021.

Chen, R., Morris, H. R., & Whitmore, P. M. (2013). Fast detection of hydrogen sulfide gasin the ppmv range with silver nanoparticle films at ambient conditions. Sensors andActuators B, Chemical, 186, 431–438.

Choudhary, M., Singh, N. K., Mishra, V. N., & Dwivedi, R. (2013). Selective detection ofhydrogen sulfide using copper oxide-doped tin oxide based thick film sensor array.Materials Chemistry and Physics, 142(1), 370–380.

Coombs, C. E. O., Holman, B. W. B., Friend, M. A., & Hopkins, D. L. (2017). Long-term redmeat preservation using chilled and frozen storage combinations: A review. MeatScience, 125(Supplement C), 84–94.

Corrêa, A. C., de Morais Teixeira, E., Pessan, L. A., & Mattoso, L. H. C. (2010). Cellulosenanofibers from curaua fibers. Cellulose, 17(6), 1183–1192.

Deepa, B., Abraham, E., Pothan, L. A., Cordeiro, N., Faria, M., & Thomas, S. (2016).Biodegradable nanocomposite films based on sodium alginate and cellulose nanofi-brils. Materials, 9(1), 50.

Dufresne, A. (2012). Nanocellulose: From nature to high performance tailored materials. DeGruyter.

Elimelech, M., Nagai, M., Ko, C.-H., & Ryan, J. N. (2000). Relative insignificance of mi-neral grain zeta potential to colloid transport in geochemically heterogeneous porousmedia. Environmental Science & Technology, 34(11), 2143–2148.

Foresti, M. L., Vázquez, A., & Boury, B. (2017). Applications of bacterial cellulose asprecursor of carbon and composites with metal oxide, metal sulfide and metal na-noparticles: A review of recent advances. Carbohydrate Polymers, 157, 447–467.

Fu, L., Zhang, J., & Yang, G. (2013). Present status and applications of bacterial cellulose-based materials for skin tissue repair. Carbohydrate Polymers, 92(2), 1432–1442.

Gahlaut, S. K., Yadav, K., Sharan, C., & Singh, J. P. (2017). Quick and selective dual modedetection of H2S gas by mobile app employing silver nanorods array. AnalyticalChemistry, 89(24), 13582–13588.

George, J., Ramana, K. V., Bawa, A. S., & Siddaramaiah (2011). Bacterial cellulose na-nocrystals exhibiting high thermal stability and their polymer nanocomposites.International Journal of Biological Macromolecules, 48(1), 50–57.

Gonçalves, L. F. F. F., Silva, C. J. R., Kanodarwala, F. K., Stride, J. A., Pereira, M. R., &Gomes, M. J. M. (2013). Synthesis and characterization of organic–inorganic hybridmaterials prepared by sol–gel and containing ZnxCd1−xS nanoparticles prepared bya colloidal method. Journal of Luminescence, 144, 203–211.

Gong, L., & Haur, S. C. (2017). On demand rapid patterning of colored amorphous mo-lybdenum oxide painted via a focused laser beam. Journal of Materials Chemistry C, 5.

Günter, J. R. (1972). Topotactic dehydration of molybdenum trioxide-hydrates. Journal ofSolid State Chemistry, 5(3), 354–359.

Infahsaeng, Y., & Ummartyotin, S. (2017). Investigation on structural properties of Al-substituted ZnS particle prepared from wet chemical synthetic route. Results inPhysics, 7, 1245–1251.

Janpetch, N., Saito, N., & Rujiravanit, R. (2016). Fabrication of bacterial cellulose-ZnOcomposite via solution plasma process for antibacterial applications. CarbohydratePolymers, 148, 335–344.

Ji, W., Shen, R., Yang, R., Yu, G., Guo, X., Peng, L., ... Ding, W. (2013). Partially nitrided


11


http://refhub.elsevier.com/S0144-8617(19)31234-2/sbref0005



















































































molybdenum trioxide with promoted performance as an anode material for lithium-ion batteries. Journal of Materials Chemistry A, 2.

Jozala, A. F., de Lencastre-Novaes, L. C., Lopes, A. M., de Carvalho Santos-Ebinuma, V.,Mazzola, P. G., Pessoa-Jr, A., ... Chaud, M. V. (2016). Bacterial nanocellulose pro-duction and application: A 10-year overview. Applied Microbiology and Biotechnology,100(5), 2063–2072.

Kabcum, S., Tammanoon, N., Wisitsoraat, A., Tuantranont, A., Phanichphant, S., &Liewhiran, C. (2016). Role of molybdenum substitutional dopants on H2S-sensingenhancement of flame-spray-made SnO2 nanoparticulate thick films. Sensors andActuators B, Chemical, 235, 678–690.

Kersen, Ü., & Holappa, L. (2006). Surface characterization and H2S-sensing potential ofiron molybdate particles produced by supercritical solvothermal method and sub-sequent oxidation. Applied Physics A, 85(4), 431–436.

Keshk, S. (2014). Bacterial cellulose production and its industrial applications. Journal ofBioprocessing & Biotechniques, 4.

Kurosumi, A., Sasaki, C., Yamashita, Y., & Nakamura, Y. (2009). Utilization of variousfruit juices as carbon source for production of bacterial cellulose by Acetobacterxylinum NBRC 13693. Carbohydrate Polymers, 76(2), 333–335.

Lee, C. M., Gu, J., Kafle, K., Catchmark, J., & Kim, S. H. (2015). Cellulose produced byGluconacetobacter xylinus strains ATCC 53524 and ATCC 23768: Pellicle formation,post-synthesis aggregation and fiber density. Carbohydrate Polymers, 133, 270–276.

Li, H.-Y., Huang, L., Wang, X.-X., Lee, C.-S., Yoon, J.-W., Zhou, J., ... Lee, J.-H. (2017).Molybdenum trioxide nanopaper as a dual gas sensor for detecting trimethylamineand hydrogen sulfide. RSC Advances, 7(7), 3680–3685.

Li, W.-J., Shi, E.-W., Zhong, W.-Z., & Yin, Z.-W. (1999). Growth mechanism and growthhabit of oxide crystals. Journal of Crystal Growth, 203(1), 186–196.

Lin, H.-M., Hsu, C.-M., Yang, H.-Y., Lee, P.-Y., & Yang, C.-C. (1994). NanocrystallineWO3-based H2S sensors. Sensors and Actuators B, Chemical, 22(1), 63–68.

Lin, N., & Dufresne, A. (2014). Nanocellulose in biomedicine: Current status and futureprospect. European Polymer Journal, 59(Supplement C), 302–325.

Lin, N., Huang, J., & Dufresne, A. (2012). Preparation, properties and applications ofpolysaccharide nanocrystals in advanced functional nanomaterials: A review.Nanoscale, 4.

MalekAlaie, M., Jahangiri, M., Rashidi, A. M., HaghighiAsl, A., & Izadi, N. (2015).Selective hydrogen sulfide (H2S) sensors based on molybdenum trioxide (MoO3)nanoparticle decorated reduced graphene oxide. Materials Science in SemiconductorProcessing, 38, 93–100.

Malone Rubright, S. L., Pearce, L. L., & Peterson, J. (2017). Environmental toxicology ofhydrogen sulfide. Nitric Oxide, 71, 1–13.

Maneerung, T., Tokura, S., & Rujiravanit, R. (2008). Impregnation of silver nanoparticlesinto bacterial cellulose for antimicrobial wound dressing. Carbohydrate Polymers,72(1), 43–51.

Martínez-Sanz, M., Lopez-Rubio, A., & Lagaron, J. M. (2011). Optimization of the na-nofabrication by acid hydrolysis of bacterial cellulose nanowhiskers. CarbohydratePolymers, 85(1), 228–236.

McMillin, K. W. (2017). Advancements in meat packaging. Meat Science, 132, 153–162.Mohebi, E., & Marquez, L. (2015). Intelligent packaging in meat industry: An overview of

existing solutions. Journal of Food Science and Technology, 52(7), 3947–3964.Murphy, K. M., O’Grady, M. N., & Kerry, J. P. (2013). Effect of varying the gas headspace

to meat ratio on the quality and shelf-life of beef steaks packaged in high oxygenmodified atmosphere packs. Meat Science, 94(4), 447–454.

Pandey, S. K., Kim, K.-H., & Tang, K.-T. (2012). A review of sensor-based methods formonitoring hydrogen sulfide. TrAC Trends in Analytical Chemistry, 32, 87–99.

Remenant, B., Jaffrès, E., Dousset, X., Pilet, M.-F., & Zagorec, M. (2015). Bacterial spoilersof food: Behavior, fitness and functional properties. Food Microbiology, 45(Part A),45–53.

Ross, P., Mayer, R., & Benziman, M. (1991). Cellulose biosynthesis and function in bac-teria. Microbiological Reviews, 55(1), 35–58.

Sarala Devi, G., Manorama, S., & Rao, V. J. (1995). High sensitivity and selectivity of anSnO2 sensor to H2S at around 100 °C. Sensors and Actuators B, Chemical, 28(1), 31–37.

Sarfraz, J., Ihalainen, P., Määttänen, A., Gulin, T., Koskela, J., Wilén, C. E., ... Peltonen, J.(2014). A printed H2S sensor with electro-optical response. Sensors and Actuators B,Chemical, 191, 821–827.

Schirmer, B. C., Heir, E., & Langsrud, S. (2009). Characterization of the bacterial spoilageflora in marinated pork products. Journal of Applied Microbiology, 106(6), 2106–2116.

Shah, N., Islam, M., Khattak, W., & Park, J. (2013). Overview of bacterial cellulosecomposites: A multipurpose advanced material. Carbohydrate Polymers, 98.

Shimizu, K., Imai, H., Hirashima, H., & Tsukuma, K. (1999). Low-temperature synthesis ofanatase thin films on glass and organic substrates by direct deposition from aqueoussolutions. Thin Solid Films, 351(1), 220–224.

Singhsa, P., Narain, R., & Manuspiya, H. (2018a). Bacterial cellulose nanocrystals (BCNC)preparation and characterization from three bacterial cellulose sources and devel-opment of functionalized BCNCs as nucleic acid delivery systems. ACS Applied NanoMaterials, 1(1), 209–221.

Singhsa, P., Narain, R., & Manuspiya, H. (2018b). Physical structure variations of bac-terial cellulose produced by different Komagataeibacter xylinus strains and carbonsources in static and agitated conditions. Cellulose, 25(3), 1571–1581.

Sofos, J. N. (1994). Microbial growth and its control in meat, poultry and fish. In A. M.Pearson, & T. R. Dutson (Eds.). Quality attributes and their measurement in meat, poultryand Fish products (pp. 359–403). Boston, MA: Springer US.

Solis, J. L., Saukko, S., Kish, L. B., Granqvist, C. G., & Lantto, V. (2001). Nanocrystallinetungsten oxide thick-films with high sensitivity to H2S at room temperature. Sensorsand Actuators B, Chemical, 77(1), 316–321.

Susi, T., Pichler, T., & Ayala, P. (2015). X-ray photoelectron spectroscopy of graphiticcarbon nanomaterials doped with heteroatoms. Beilstein Journal of Nanotechnology, 6,177–192.

Szymańska-Chargot, M., Cybulska, J., & Zdunek, A. (2011). Sensing the structural dif-ferences in cellulose from apple and bacterial cell wall materials by Raman and FT-IRspectroscopy. Sensors (Basel, Switzerland), 11(6), 5543–5560.

Tang, J., Sisler, J., Grishkewich, N., & Tam, K. C. (2017). Functionalization of cellulosenanocrystals for advanced applications. Journal of Colloid and Interface Science, 494,397–409.

Ummartyotin, S., & Manuspiya, H. (2015). A critical review on cellulose: From funda-mental to an approach on sensor technology. Renewable and Sustainable EnergyReviews, 41(Supplement C), 402–412.

Vasconcelos, N. F., Feitosa, J. P. A., da Gama, F. M. P., Morais, J. P. S., Andrade, F. K., deSouza Filho, M. S. M., & Rosa, M. F. (2017). Bacterial cellulose nanocrystals producedunder different hydrolysis conditions: Properties and morphological features.Carbohydrate Polymers, 155(Supplement C), 425–431.

Wada, M., Kondo, T., & Okano, T. (2003). Thermally induced crystal transformation fromcellulose I to Iβ. Polymer Journal, 35, 155–159.

Wada, M., Okano, T., & Sugiyama, J. (2001). Allomorphs of native crystalline cellulose Ievaluated by two equatoriald-spacings. Journal of Wood Science, 47(2), 124–128.

Wang, C., Chu, X., & Wu, M. (2006). Detection of H2S down to ppb levels at roomtemperature using sensors based on ZnO nanorods. Sensors and Actuators B, Chemical,113(1), 320–323.

Winter, H. T., Cerclier, C., Delorme, N., Bizot, H., Quemener, B., & Cathala, B. (2010).Improved colloidal stability of bacterial cellulose nanocrystal suspensions for theelaboration of spin-coated cellulose-based model surfaces. Biomacromolecules, 11(11),3144–3151.

Wu, J.-M., & Liu, R.-H. (2012). Thin stillage supplementation greatly enhances bacterialcellulose production by Gluconacetobacter xylinus. Carbohydrate Polymers, 90(1),116–121.

Xu, J., Wang, X., & Shen, J. (2006). Hydrothermal synthesis of In2O3 for detecting H2S inair. Sensors and Actuators B, Chemical, 115(2), 642–646.

Yamamoto, H., Horii, F., & Hirai, A. (1996). In situ crystallization of bacterial cellulose II.Influences of different polymeric additives on the formation of celluloses Iα and Iβ atthe early stage of incubation. Cellulose, 3(1), 229–242.

Yang, X., Zhu, Z., Liu, Q., Chen, X., & Ma, M. (2008). Effects of PVA, agar contents, andirradiation doses on properties of PVA/ws-chitosan/glycerol hydrogels made by γ-irradiation followed by freeze-thawing. Radiation Physics and Chemistry, 77(8),954–960.

Zhang, G., Liao, Q., Zhang, Z., Liang, Q., Zhao, Y., Zheng, X., ... Zhang, Y. (2016). Novelpiezoelectric paper-based flexible nanogenerators composed of BaTiO3 nanoparticlesand bacterial cellulose. Advanced Science, 3(2) 1500257-n/a.


12






























































































































fabrication of hybrid thin film based on bacterial

Documents